Perl 6 - the future is here, just unevenly distributed

IRC log for #salt, 2013-12-19

| Channels | #salt index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:03 modafinil dwfreed: my fault :3 doing silly things with job config files
00:03 modafinil cheers
00:03 dwfreed heh
00:10 Sheco joined #salt
00:11 jacksontj joined #salt
00:11 xmltok_ joined #salt
00:19 shadfc joined #salt
00:20 cnelsonsic joined #salt
00:22 bhosmer joined #salt
00:22 dstanek joined #salt
00:27 shadfc question: https://gist.github.com/jwineinger/8032282. this was creating the statsd home directory with bad permissions so that the git command could not clone into it. How would I prevent that?  are there mode and owner require that I can use?
00:28 shadfc or how can i reference as a require the cfg below where i specifically set the mode, owner, group on that home directory?
00:28 xmltok joined #salt
00:30 xinkeT joined #salt
00:41 dstanek joined #salt
00:45 th3reverend joined #salt
00:46 jslatts joined #salt
00:48 Sheco joined #salt
00:50 higgs001 joined #salt
00:57 th3reverend left #salt
00:57 fllr joined #salt
01:01 fllr joined #salt
01:08 filecopy joined #salt
01:08 filecopy question about file.copy
01:08 filecopy well, acutally more general question about file serving
01:08 filecopy I want to copy a file to the host only if it doesn't exist already
01:09 filecopy would the best way to do that be file.managed with an unless: ls /path/to/file ?
01:09 filecopy file.copy only works on the minion from path to path is my understanding from the source code
01:09 shadfc i just did this with something else:  - unless: test -f /path/to/file
01:11 filecopy ah, I didn't RTFM :-[  "replace If this file should be replaced. If false, this command will not overwrite file contents but will enforce permissions if the file exists already. Default is True."
01:11 filecopy that's under file.managed
01:12 filecopy ofc i needed to come on irc before reading the docs
01:12 filecopy thanks though shadfc !
01:12 shadfc np
01:12 filecopy saltstack rules, puppchefsible drools
01:23 cachedout joined #salt
01:25 mgw joined #salt
01:26 bemehow_ joined #salt
01:27 smccarthy joined #salt
01:29 troyready joined #salt
01:36 Nexpro joined #salt
01:38 anitak joined #salt
01:42 jY- is there a way in salt for a minion on a load balancer to say i want to get all web#.domain.com or if I make a grain like server_type that equals web
01:42 jY- where i can pull all the web type servers so i can build a list of servers to put in haproxy's template file for backend
01:45 fllr joined #salt
01:45 wonhunawks joined #salt
01:47 mayfield joined #salt
01:53 rojem joined #salt
01:57 frosty996 jY-: you could use the salt mine
01:59 ajw0100 joined #salt
02:02 oz_akan_ joined #salt
02:13 sroegner joined #salt
02:15 sroegner_ joined #salt
02:20 favadi joined #salt
02:23 bhosmer joined #salt
02:24 bemehow joined #salt
02:26 meteorfox joined #salt
02:26 favadi joined #salt
02:27 bemehow_ joined #salt
02:31 meteorfox joined #salt
02:34 DanGarthwaite joined #salt
02:38 capricorn_1 joined #salt
02:38 anitak joined #salt
02:42 jslatts joined #salt
02:56 meteorfo_ joined #salt
02:58 xl1 joined #salt
02:59 halfss__ joined #salt
03:00 wonhunawks joined #salt
03:01 jfzhu_us joined #salt
03:16 mgw joined #salt
03:20 lyddonb joined #salt
03:22 jalbretsen joined #salt
03:24 MTecknology Damnit... my reactor system stopped working
03:27 rgarcia_ joined #salt
03:27 jergerber joined #salt
03:32 ajw0100 joined #salt
03:35 higgs001 joined #salt
03:37 jergerber joined #salt
03:39 anitak joined #salt
03:39 MTecknology Did the reactor system do a lot of changing or what the crap?
03:40 MTecknology This doesn't seem to work anymore -->  {% if not data['result'] and data['id'].startswith('boothost') and data['id'] != 'boothost' %}
03:40 jfzhu_us I am trying to make the existence of a user's home directory a requisite via the following syntax:
03:40 jfzhu_us - require:
03:40 jfzhu_us - file.directory:
03:40 jfzhu_us - name: /home/jzhu
03:40 jfzhu_us but I am getting illegal requisite and i am not sure why
03:41 MTecknology use a pastebin next time, you want   -require: \n\t -file: /home/jzhu
03:44 oz_akan_ joined #salt
03:47 anuvrat joined #salt
03:52 tomtomtomtom joined #salt
03:53 MTecknology running   salt-minion -l debug  shows me   [INFO    ] Public key rejected for boothost.corp.good-sam.com   [DEBUG   ] Gathering reactors for tag salt/auth  || in /etc/salt/master I have reactor: \n\t - 'auth': \n - /srv/reactor/auth-pending.sls
04:00 jfzhu_us With regards to requisites. Do you need a state file to correspond with it?
04:02 vipul joined #salt
04:04 MTecknology damnit... apparently all of the tags changed
04:05 aranhoide joined #salt
04:05 MTecknology jfzhu_us: a state, yes
04:05 aranhoide does 0.17.4 include the merged salt-cloud, or hasn't that made it into a release yet?
04:06 jimallman joined #salt
04:09 jfzhu_us MTecknology: thanks
04:11 MTecknology damnit... I can't tolerate these changes right now
04:11 MTecknology time to go home
04:14 shadowsun forresta: you around?
04:15 dstanek joined #salt
04:16 EWDurbin joined #salt
04:18 favadi joined #salt
04:19 shadowsun basepi: you around?
04:23 MTecknology shadowsun: betcha they're asleep
04:23 pdayton joined #salt
04:23 shadowsun MTecknology: probably right
04:23 shadowsun I'm just trying to wrap my head around salt + gitfs + branches/environments
04:24 bhosmer joined #salt
04:24 shadowsun because everything is fine right up until top.sls can only exist in the master branch
04:24 shadowsun at which point I'm going "Wait a minute now I can't merge my dev branch back to master... erm... " and then my brain spits out sparks.
04:25 shadowsun I'm having very little luck finding information on the internets about how people have actually implemented this... so i figured I'd hope someone might be around who knew.
04:25 shadowsun do you know?
04:25 MTecknology I'm trying to figure out how to re-implement the reactor system
04:25 MTecknology never used gittfs
04:26 shadowsun Haven't used the reactor system yet
04:26 shadowsun otherwise I'd volunteer
04:26 MTecknology the tags completely changed :(
04:26 rgarcia_ joined #salt
04:26 MTecknology the docs lack explaining how to figure out what tag you need
04:27 shadowsun argh
04:27 shadowsun I know your pain!
04:27 shadowsun I felt it when batch mode magically started inheriting default timeout but this wasn't mentioned in the docs
04:28 shadowsun I almost crashed a couple nodes
04:32 MTecknology I love how fast salt progresses, but some of these changes are a giant bitch
04:32 shadowsun Yush
04:32 shadowsun I had to manually apply a patch to all my minions today because of a regression in 0.17.2
04:32 shadowsun which is what's still in EPEL
04:32 MTecknology I don't want to have to review change logs and commit logs to figure out if a newer version will completely break things :(
04:32 shadowsun haha yeah
04:32 shadowsun I'm about to start rolling my own RPMs
04:32 MTecknology I absolutely hate rpm based distros
04:33 shadowsun We've already got a build server with mock and everything and I've already got a repo...
04:33 shadowsun ...YES>
04:33 shadowsun I'm actually a Slackware guy
04:33 ConceitedCode joined #salt
04:33 shadowsun But at work it's all CentOS.. so.. I make CentOS work.. but ugh
04:33 shadowsun RPMs are such.. ick.
04:33 shadowsun I can see how the original design was kind of neat
04:33 shadowsun but then it's like.... lets play pin the tail on the donkey for ten years and see what we get
04:36 MTecknology I've been purging rhel from my company
04:36 shadowsun Yeah... not gonna happen here
04:36 MTecknology in about 1yr, we should have ~7 centos boxes remaining
04:36 shadowsun How many boxes did you start with?
04:37 shadowsun if you can even tell me that
04:37 MTecknology 40
04:37 shadowsun that's not bad at all
04:37 MTecknology they were all rhel, within 1yr of me starting, they became centos boxes
04:37 shadowsun Yeah
04:37 MTecknology if vendors require rhel, then I put on centos; otherwise it's debian
04:37 shadowsun because cheaper for most of the same thing
04:38 shadowsun My experience with Debian was apparently back in the dark days
04:38 MTecknology it's not like you actually pay for support... you pay for updates
04:38 shadowsun I'm told it's much better now
04:38 MTecknology I absolutely LOVE debian
04:38 Jahkeup I'm coming to love vanilla debian too
04:38 MTecknology looooooove
04:38 shadowsun I'm going to give it another try one of these days
04:38 shadowsun Inside a KVM
04:38 shadowsun where it's safe
04:39 MTecknology what do you use for virtualization?
04:39 shadowsun KVM
04:39 MTecknology look into proxmox
04:39 shadowsun i.e. I don't even libvirt
04:39 shadowsun I just run qemu-binaries
04:39 shadowsun It's on my list, actually, of ones to look into
04:39 MTecknology proxmox is fucking amazing
04:39 shadowsun Since we'll probably need a frontend at some point for others
04:40 shadowsun And tbh because it looks shiny
04:40 MTecknology at home, I can set up dns for a server, deploy the server, have it configured in salt, and state.highstate configured 100% in <3min
04:40 shadowsun How easy is it to mannually customize the flags passed to qemu-system with it?
04:40 MTecknology It actually takes ~5min when I'm doing it casually
04:40 shadowsun Yep
04:41 shadowsun My initial highstate takes about 20 minutes
04:41 shadowsun But, there's cPanel ... soooo things get recompiled when I run my initial highstate
04:41 MTecknology at work, 180 of the boxes will take ~40min the first time and ~5min after that
04:41 shadowsun Yeah
04:42 shadowsun Mine are usually less for runs after that, until I trigger a new easyapache build to update httpd/php/etc everywhere
04:42 MTecknology I've only ever needed to change what kernel options are being sent to the vm when running openvz containers and that was dead simple (for ipv6)
04:42 shadowsun Yeah
04:42 shadowsun but that's ovz
04:42 MTecknology stop using apache!
04:42 shadowsun For Nginx? no
04:42 shadowsun Apache does what I need better
04:42 MTecknology *cough*bullshit*cough*
04:43 * shadowsun shrugs
04:43 shadowsun there's not much static content
04:43 MTecknology I really truly love nginx+php-fpm
04:43 shadowsun yes, so do I
04:43 shadowsun But our environment isn't going to handle that very well
04:44 MTecknology apache+php-fpm would be ok too, but would be a waste when nginx would be easier; apache+mod_php is a royal pile of shit
04:44 shadowsun I don't use either
04:44 MTecknology mod_php is something even apache devs claim is horrible scum
04:44 shadowsun yes
04:44 shadowsun I haven't used dso in years
04:44 MTecknology what is it that you use?
04:44 shadowsun php-fpm is awesome unless you have a ton of different users, at which point things start to get tricky
04:45 shadowsun (different users + different vhosts)
04:45 shadowsun mod_event + suphp/suexec
04:45 MTecknology nice
04:45 shadowsun Yes
04:45 shadowsun All kinds of problems I don't have with that
04:46 shadowsun Ooo
04:46 shadowsun gran turismo six finished downloading
04:47 MTecknology when I was doing shared hosting I had nginx configured so if I wanted to host a drupal site, I could just do  server { server_name domain.tld; include apps/drupal6; } and the php stuff for that site was already set up and php-fpm had its own pool for that site and that site was also in a chroot, any devs connecting to work on that site did so inside of a chroot
04:47 shadowsun never mind, another download
04:47 shadowsun That's actually not bad at all
04:48 shadowsun cPanel does its own chroot-like thing it calls jailshell
04:48 shadowsun But the problem with php-fpm pool is that a shared environment may have hundreds or thousands of users
04:48 shadowsun too much RAM
04:48 shadowsun not enough chips
04:48 shadowsun server gets sad fast.
04:49 MTecknology static two children with 32MB ram (unles more was needed)
04:49 shadowsun Depending on the site php may eat a lot more per child
04:49 ravibhure joined #salt
04:49 MTecknology the big thing with php-fpm is that people leave it wyth dynamic children; you end up with the performance (almost) of mod_php
04:49 shadowsun I've seen ~200 thanks to magento
04:50 MTecknology I had one site that would eat 200+; it was a drupal site with a few hundred modules added
04:50 shadowsun I've always locked my php-fpm pools down based off of threads and cores, but on those servers I'm not using php-fpm, i'm using the new proxy modules out of apache so I get connection queuing
04:50 MTecknology that site still isn't up...
04:50 shadowsun wait
04:50 shadowsun I'm tired
04:50 shadowsun that came out wrong.
04:50 shadowsun fcgi.
04:51 ajw0100 joined #salt
04:51 MTecknology thelegaldeal.com <-- My company was contracted to build that site, some ass hats decided to steal that company from me.
04:51 shadowsun >.<
04:51 MTecknology As long as I never step foot in web hosting/development again, I'm safe from law suites
04:52 vlcn left #salt
04:52 MTecknology it was supposed to be done in 1yr, that deadline was 3.5yr ago
04:52 shadowsun O.o
04:53 MTecknology beginr.com <-- the "new" company that took that contract
04:53 shadowsun Oh.
04:53 shadowsun OH.
04:53 shadowsun My eyes.
04:53 shadowsun they hurt.
04:53 shadowsun Negative spaces in all the wrong places.
04:53 MTecknology :P
04:54 MTecknology they're somehow still sticking to not having to go find real jobs
04:54 ravibhure joined #salt
04:55 jfzhu_us joined #salt
04:56 shadowsun :|
04:56 oz_akan_ joined #salt
05:03 psyl0n joined #salt
05:09 bhosmer joined #salt
05:18 elithrar joined #salt
05:19 sroegner_ joined #salt
05:20 pcurry_nomi joined #salt
05:31 gmoro_ joined #salt
05:51 matanya joined #salt
06:02 malinoff joined #salt
06:11 anuvrat joined #salt
06:12 Destro joined #salt
06:12 forresta shadowsun I am now
06:12 forresta Granted I plan on going to sleep soon :P
06:13 sunand joined #salt
06:25 bhosmer joined #salt
06:29 forresta shadowsun, I've gotta head to bed and I'm too tired to read the scrollback. Let me know if you got it figured out tomorrow.
06:34 anuvrat joined #salt
06:57 dstanek joined #salt
07:07 santosh_kumar joined #salt
07:07 santosh_kumar Hell Guys
07:08 santosh_kumar i want to setup the saltstack on ubuntu 12.04 server
07:11 MTecknology k?
07:13 matanya joined #salt
07:35 ajw0100 joined #salt
07:40 stefanmonkey joined #salt
07:42 stefanmonkey I run the cmd like 'salt o-manage201 grains.item ip_interfaces' , and it return the result like 'ip_interfaces: {'lo': ['127.0.0.1'], 'eth0': ['192.168.1.201']}' . THe question is how do i get the only eth0 value
07:42 stefanmonkey with salt cmd
07:45 sunand1 joined #salt
07:57 sunand grains["ip_interfaces"]["eth0"][0]
07:59 harobed joined #salt
08:02 rojem joined #salt
08:03 balboah joined #salt
08:07 fllr joined #salt
08:08 aleszoulek joined #salt
08:10 fllr joined #salt
08:13 Sheco joined #salt
08:14 stefanmonkey It seems not work
08:14 rcb joined #salt
08:14 stefanmonkey did I run wrong cmd?
08:22 elsmorian joined #salt
08:26 bhosmer joined #salt
08:27 gammalget joined #salt
08:27 rcb @stefanmonkey: salt o-manage201 grains.get ip_interfaces:eth0
08:28 stefanmonkey oh  tks a lot
08:30 stefanmonkey the doc about grains should add some stuff like this example
08:31 rcb http://salt.readthedocs.org/en/latest/ref/modules/all/salt.modules.grains.html?highlight=grains#salt.modules.grains.get
08:32 rcb Perhaps enriching the "CLI Example" on that section? I think you can just insert you example below that, it would be a nice addition :)
08:40 gammalget joined #salt
08:42 giantlock_ joined #salt
08:42 anitak joined #salt
08:44 junedm joined #salt
08:50 netzmonster joined #salt
08:50 helminthe joined #salt
08:52 dstanek joined #salt
08:56 jpihl joined #salt
09:08 fllr joined #salt
09:10 druonysus joined #salt
09:12 gammalget joined #salt
09:16 krak3n` joined #salt
09:28 junedm left #salt
09:28 tomtomtomtom joined #salt
09:29 wonhunawks joined #salt
09:29 ossalt joined #salt
09:32 N-Mi joined #salt
09:32 N-Mi joined #salt
09:41 carlos joined #salt
09:44 JasonG_TA joined #salt
09:48 MrTango joined #salt
09:49 cryptomnesia joined #salt
09:53 Sheco joined #salt
09:57 joehoyle joined #salt
10:04 luminous is there a way to 'list all available salt module/functions' ?
10:04 luminous eg, is there a module.function somewhere that does this?
10:05 luminous I'm looking but not seeing, and want to confirm I'm not overlooking
10:08 fllr joined #salt
10:09 bhosmer joined #salt
10:09 Sheco joined #salt
10:10 elfixit joined #salt
10:11 rcb @luminous, does "salt '*' -d" gives what you need? (check the salt man page for info on the -d switch)
10:13 gammalget joined #salt
10:19 luminous rcb: no, that's not what I had in mind
10:20 luminous rcb: similar to grains.ls, I'm thinking of a modules.ls
10:20 luminous or modules.ls_func
10:20 luminous or whatever
10:20 luminous but a way to list modules and functions salt sees
10:24 gammalget joined #salt
10:25 ahale sys.list_functions and sys.list_modules ?
10:26 bhosmer_ joined #salt
10:34 sunand joined #salt
10:39 TheSojourner joined #salt
10:42 aarontc joined #salt
10:42 gammalget joined #salt
10:43 anitak joined #salt
10:44 EugeneKay joined #salt
10:53 dstanek joined #salt
10:55 gammalget joined #salt
10:56 jY- is there a way in salt for a minion on a load balancer to say i want to get all web#.domain.com or if I make a grain like server_type that equals web
10:56 jY- where i can pull all the web type servers so i can build a list of servers to put in haproxy's template file for backend
11:00 scott_w joined #salt
11:08 fllr joined #salt
11:11 krak3n` jY- you could use Salt Mine
11:11 jY- krak3n`: ok i'll take a look
11:12 krak3n` the mine can collect ip address of all your minions, you can then use that in templates to get the ip address of minions matching web#.domain.com
11:12 krak3n` http://stackoverflow.com/questions/17158665/how-to-get-a-list-of-all-salt-minions-in-a-template for example
11:13 anitak joined #salt
11:23 diegows joined #salt
11:24 viq joined #salt
11:29 anitak joined #salt
11:33 bhosmer joined #salt
11:34 elfixit joined #salt
11:53 pniederw joined #salt
11:53 pniederw which release is hydrogen?
12:07 higgs001 joined #salt
12:08 fllr joined #salt
12:16 sgflt joined #salt
12:18 sgflt is there a way to also install AUR packages on arch using salt?
12:24 JasonG_TA joined #salt
12:25 viq pniederw: 0.18
12:26 viq sgflt: kinda, I was playing with that, and found it easiest to first install packer, and then issue commands to that from what I remember
12:26 pcurry_nomi joined #salt
12:26 viq pniederw: I may have it gisted somewhere, let me see
12:27 viq sgflt: kinda cheating, but https://gist.github.com/viq/7369690
12:27 bhosmer joined #salt
12:28 viq And while I'm here, a syndic question: I have master, couple nodes connected to it, one of them is a syndic with one node connected to it. Should something like "salt '*' test.ping" make the node connected to syndic respond as well?
12:34 joehoyle joined #salt
12:35 sgflt viq, aur.sh has been very good to me in this regard, as well as pacaur. but thanks for the heads-up. i guess i will have a look at the salt source and try to implement a module
12:35 viq sgflt: I didn't dive too deeply into this, that's a result of me spending an afternoon to see what I can do in that regard
12:36 ajw0100 joined #salt
12:41 omame jcockhren: found the issue: the command was taking longer than the LocalClient timeout to run, so it returned {} instead of waiting for the minion to come back with the results
12:42 omame thanks for helping me
12:46 ossalt joined #salt
12:47 ossalt {% if grains['fqdn_ip4'] == '172.31.7.3' %} htmlfile: salt://apache/index1.html {% else %} htmlfile: salt://apache/index3.html {% endif %}
12:48 ossalt salt '*' grains.item fqdn_ip4
12:48 ossalt fqdn_ip4:       172.31.7.3
12:48 viq hmm, https://github.com/saltstack/salt/issues/6207
12:48 ossalt doesn't match
12:48 ossalt with the ip address
12:49 ossalt it returns index3.html
12:49 viq ossalt: try {% if '172.31.7.3' in grains['fqdn_ip4'] %}
12:50 ossalt trying
12:50 ossalt tnx it worked
12:50 ossalt what is the problem with my code?
12:51 viq fqdn_ip4 is a list, not a string, I think
12:51 ossalt hmm ok
12:51 ossalt tnx again
12:53 viq basepi: since I saw you being active on that bug, ping ;) https://github.com/saltstack/salt/issues/6207
12:55 dstanek joined #salt
13:00 viq Oh, I see there's more issues around that
13:00 troyready joined #salt
13:01 fllr joined #salt
13:02 troyready joined #salt
13:08 MrTango joined #salt
13:08 fllr joined #salt
13:09 jslatts joined #salt
13:10 bhosmer joined #salt
13:17 Whissi joined #salt
13:17 elfixit joined #salt
13:18 dstanek joined #salt
13:26 ravibhure1 joined #salt
13:28 junedm1 joined #salt
13:29 junedm1 left #salt
13:36 mortis_ anyone know when 0.17.4 will be in the apt-repo?
13:40 brianhicks joined #salt
13:41 vejdmn joined #salt
13:47 aleszoulek joined #salt
13:52 tyler joined #salt
13:58 favadi joined #salt
14:02 mpanetta joined #salt
14:03 juicer2 joined #salt
14:05 bhosmer joined #salt
14:06 bhosmer_ joined #salt
14:08 oz_akan_ joined #salt
14:08 fllr joined #salt
14:10 racooper joined #salt
14:11 psyl0n joined #salt
14:15 MK_FG joined #salt
14:17 JasonG_TA joined #salt
14:18 jslatts joined #salt
14:22 foxx joined #salt
14:23 sroegner_ joined #salt
14:23 pass_by_value joined #salt
14:25 luminous mortis_: no, no one usually knows that better than 'soon'
14:26 zwevans joined #salt
14:27 mortis_ luminous: so no auto-building of packages?
14:27 oz_akan_ joined #salt
14:28 luminous mortis_: I don't know about the ubuntu/deb release/packaging process
14:34 rojem joined #salt
14:37 fllr joined #salt
14:46 ashtonian joined #salt
14:54 kuL4 joined #salt
14:55 th3reverend joined #salt
14:55 anuvrat joined #salt
14:56 MK_FG joined #salt
15:04 timoguin_ joined #salt
15:07 [diecast] joined #salt
15:11 jergerber joined #salt
15:13 [diecast] joined #salt
15:14 kaptk2 joined #salt
15:14 JasonSwindle joined #salt
15:19 micw joined #salt
15:19 micw hi
15:19 rojem joined #salt
15:19 micw i try to get started with salt. i installed a master and a windows minion. i can ping it with salt '*' test.ping
15:20 micw now i added the default "win_gitrepos" to the master and restarted it
15:20 rojem joined #salt
15:20 micw salt '*' pkg.available_version 7zip
15:20 micw shows nothing
15:20 micw do i have to change something on my windows node?
15:22 bhosmer joined #salt
15:23 blee joined #salt
15:24 micw ah, i forgot the update ^^ found the right place in the docs now
15:28 krak3n` joined #salt
15:31 forresta joined #salt
15:32 cnelsonsic joined #salt
15:33 tyler joined #salt
15:33 fllr joined #salt
15:34 bt joined #salt
15:34 tyler-baker joined #salt
15:34 tyler-baker joined #salt
15:35 Whissi How do I access "level1" of my pillar data? -> http://pastebin.com/raw.php?i=Y30JiARp
15:36 opapo joined #salt
15:36 forresta Whissi, http://docs.saltstack.com/topics/pillar/#pillar-get-function
15:37 jesusaurus Whissi: level0 is a list of ['data', 'level1']
15:37 jesusaurus you need to index into that array
15:37 matanya joined #salt
15:38 Whissi I cannot do that from cli, can I?
15:38 MTecknology In a previous version of salt, I had two reactor tags: 'auth' and 'minion_start'. In this version, 'auth' seems to have turned to 'salt/auth' and 'minion_start' seems to be gone. Where did that tag go and what else changed? What does data look like now? Where the heck do I find that documentation?
15:39 jesusaurus Whissi: level0:1:level1
15:40 Whissi jesusaurus: Doesn't work :(
15:42 jesusaurus oh, wow, it looks like the code around pillar.get has changed since i last looked at it. it assumes dicts
15:42 jesusaurus Whissi: can you open a github issue?
15:43 jesusaurus something along the lines of "pillar.get can't index into arrays" with your example
15:44 aleszoulek joined #salt
15:44 Whissi OK, I'll open an issue
15:48 davet joined #salt
15:50 MTecknology FRICK! The reactor system is irritating!
15:53 robawt jesusaurus: you have the best name ever
15:54 jslatts joined #salt
15:54 jesusaurus :)
15:55 th3reverend left #salt
15:56 Teknix joined #salt
15:57 micw how can i do the following with saltstack: run a command that checks if a certain ca-certificate is installed in current java version. if not download a certificate file and run a command that installs that certificate
15:57 micw do i need to code a module for such?
15:58 MTecknology It seems like events are happening without a tag or without going through the reactor system... this is pissing me off
15:58 defunctzombie any examples on requiring multiple packages in a -require ?
15:59 alunduil joined #salt
16:00 MTecknology defunctzombie: http://dpaste.com/1512957/
16:00 defunctzombie MTecknology: thanks!
16:03 malinoff joined #salt
16:04 scuwolf joined #salt
16:04 rcb micw, u probably will have to write a state file. Check this part of the walkthrough: http://docs.saltstack.com/topics/tutorials/starting_states.html
16:05 elsmorian1 joined #salt
16:06 colinbits joined #salt
16:06 Nazzy joined #salt
16:06 micw rcb, that's what I finally want. I already read this section. but the statefile "describes the destination, not the journey" - so at which place do i code what has to happen?
16:07 cachedout_home joined #salt
16:07 micw my state will be something like: java.certificate.pressent: javapath: somepath, certificate: somecerturl
16:07 micw so it tells what has to be
16:07 Ahlee oh bother. anybody remember if 0.15.x had issues communicating with 0.16 masters?
16:09 forresta Ahlee, that shouldn't be an issue.
16:09 forresta it's only a 0.17+ master with 0.16.x machines or lower
16:09 nahamu I'm pretty sure that <0.17.1 minions can't talk to >=0.17.1 masters
16:09 Ahlee various levels of talk
16:09 Ahlee some things work, some don't
16:10 nahamu http://docs.saltstack.com/topics/releases/0.17.1.html
16:10 nahamu "THIS RELEASE IS NOT COMPATIBLE WITH PREVIOUS VERSIONS. If you update your master to 0.17.1, you must update your minions as well. Sorry for the inconvenience -- this is a result of one of the security fixes listed below."
16:10 ckao joined #salt
16:10 nahamu Ahlee: what version is your master running?
16:10 Ahlee yeah
16:10 Ahlee 0.16.3 in this env
16:10 Ahlee 0.15.1 minion
16:11 forresta that should be ok
16:11 nahamu I am useless to you. :)
16:11 rcb micw, check the example of making sure the file /etc/ssh/sshd_config is present. U could adapt that to have your certificate file managed like that.
16:11 Ahlee yeah, weird bug with _modules
16:11 Ahlee modules sync'd, I see them on the server
16:11 Ahlee calls to module fail saying not avial, nothing in debug log
16:11 cachedout joined #salt
16:12 micw rcb, can i combine "builtin state modules" to a new "state module"? i could do what i want by combining file.managed + cmd.execute + cmd.unless
16:12 micw but that would only work for one certificate, i had to copy all the stuff for each cert
16:13 rcb I have some time. Let's do this together in steps. Which is the complete path of the file you need to make sure exists on the minion?
16:14 smccarthy joined #salt
16:14 rcb micw, you can manage a whole dir with subdirs and files in a single state file, so if all your certs are rooted in a single dir tree, you could do that, I guess?
16:14 micw rcb: i have only 15 minutes left today, then later again ^^ but let's start
16:15 ksk joined #salt
16:15 ksk hey guyse
16:15 MTecknology dangit.... my issue was a bug in salt
16:17 micw rcb: a certificate is an entry in a certificate database
16:17 micw i can run a command to check if it exists:
16:18 micw keytool.exe -list -keystore /lib/security/cacerts -alias "MyCert"
16:18 micw needs to be executed in $JAVA_DIR
16:18 micw and exits with 0 if the certifiacte exists
16:18 micw so that's my "unless" part of the command
16:19 foxx joined #salt
16:19 micw keytool.exe  -importcert -alias "MyCert"  -file /path/to/cert -keystore lib/security/cacerts (...) has to be called if the cert not exists
16:19 MTecknology a bug... and all sorts of stuff changing
16:19 rcb micw, this (http://dpaste.com/1512960/) will be a state failure if the cert doesn't exist.
16:19 foxx uh -V rather
16:20 foxx oops, wrong channel
16:20 colinbits joined #salt
16:20 rcb micw, working... please wait...
16:20 micw rcb: great that goes in the right direction!
16:22 rcb micw, this (http://dpaste.com/1512963/) has the unless clause and if that is not satisfied, will run the install cmd
16:22 ksk hey guyse
16:23 ksk i know how to use "cp.get_dir" in a salt-call; however i dont get howto use it in a salt-module. i did something like "__salt__['cp.get_dir']('salt://files/foo/', '/bar')" - but nothing happens
16:23 ksk how would i do that / whats my mistake?
16:25 micw rcb: ok so far. but now i need to "call" this sls for several certs, probably with different $JAVA_DIR
16:25 quickdry21 joined #salt
16:25 penguin_dan joined #salt
16:27 rcb micw, I am sure that could be done using pillar. But I am yet to study that myself. You probably want to declare a list of tuples in pillar ( each entry being JAVA_DIR, cert_name ) and then use that on the state file, inside a loop. Sorry but I can't help w/ the syntax cause I do not know it yet.
16:27 micw i see. but thank you!
16:28 viq micw: / rcb https://github.com/viq/cm-lab-salt/blob/master/salt/roots/salt/users/group1.sls
16:28 viq Or jesusaurus's example: https://github.com/jesusaurus/hpcs-salt-state/blob/master/apt/sample.pillar https://raw.github.com/jesusaurus/hpcs-salt-state/master/apt/repo/distro
16:29 micw can i use my sls in a different sls?
16:29 micw e.g.:
16:29 micw mycert:
16:29 micw cert_install:
16:29 micw ...
16:29 micw and pass "variables" to it?
16:30 viq micw: 'include' and 'context' are keywords you want
16:31 micw viq, but in this case i only could "extend" it and cannot define own properties, right?
16:31 pcurry_nomi joined #salt
16:31 min2 joined #salt
16:32 viq There's mention of extending as well. I'm not sure what you want to do
16:32 micw viq: with that example: http://dpaste.com/1512963/
16:32 pcurry_nomi1 joined #salt
16:33 viq sounds exactly like 'context', let me find it
16:33 micw viq: i want to use that like this:
16:33 micw http://dpaste.com/1512977/
16:33 micw and use "certname", "certpath" and "java_dir" within the example
16:35 scuwolf left #salt
16:35 min2 Has someone successfully have Salt-minion + Salt-master 0.14.0  working in Solaris ?
16:36 micw sorry, i need to leave now. i'll let the chat open and be back in about 2-3h
16:36 micw thanks a lot!
16:36 psyl0n joined #salt
16:36 min2 I don't have issue with salt-minion 0.14.0 on Solaris. but the master won't start up. I got: /var/opt/csw/svc/method/svc-salt-master: /etc/opt/csw/salt: bad number
16:37 min2 I use default master config file.
16:37 min2 Then I thought, let me use Salt-master from Ubuntu to control the Salt-minion of Solaris.
16:38 Whissi_ joined #salt
16:38 min2 but since the salt didn't work cross version,  the salt-master on Ubuntu  0.17.2 won't be able to communicate with salt-minion 0.14.0  on Solaris.
16:38 viq micw: I probably would do that with pillars and looping over them, like done in the links i posted above
16:38 rcb viq, micw, what about this http://dpaste.com/1512990/
16:39 viq rcb: no, pillar is "key, value", not "var1, var2, var3"
16:39 min2 I could not find  salt-master 0.14.0  for Ubuntu. So I'm stuck.
16:39 viq min2: maybe http://docs.saltstack.com/topics/tutorials/esky.html will be of use?
16:43 renoirb I have a question regarding require:
16:44 renoirb What happens if I have many requires: (with a S) instead of the singular?
16:44 junedm joined #salt
16:44 min2 vic: thanks. I'm reading  it now.
16:44 viq renoirb: I think it's considered a typo
16:44 renoirb oof.
16:45 renoirb I realize a lot of files has it
16:45 renoirb Generally it's a list right under.
16:46 viq renoirb: don't take my word for it though, I'm no expert
16:46 renoirb ok
16:46 renoirb Let's see if it breaks :[
16:46 rcb renoirb, I have written states with several requires like this: """ - require: \n - cmd: check_prereq \n - pkg: prereq_pkg \n - file: /some/needed/file"""
16:46 renoirb Me too. But several other has 'requires:' with the same type of list
16:47 renoirb I did not start the project and when I started creating my own I followed. But right at this moment I was reading the http://docs.saltstack.com/ref/states/ordering.html and got surprised to see no mention of the singular form.
16:48 viq http://docs.saltstack.com/ref/states/requisites.html doesn't show 'requires'
16:48 renoirb .. sorry, plural form that was not used.
16:48 renoirb i know
16:48 renoirb that's why im asking.
16:49 renoirb And ordering seemed to work even with that
16:49 renoirb i'm about to test if it changes the behavior when I apply with the singular form.
16:50 forresta renoirb, you don't use requires, it's always just require
16:51 forresta Pretty sure that requires: doesn't even work
16:51 renoirb forresta: Many states made by Ryan was actually using 'requires'... with an S.
16:51 forresta really??
16:51 forresta I could have sworn that generated an error
16:51 renoirb Me too...
16:51 renoirb yes.
16:52 CheKoLyN joined #salt
16:52 renoirb Sad that I am halfway through removing private data... otherwise I would give you the github repo.
16:52 renoirb but it's not ready yet
16:52 forresta all good
16:53 rcb renoirb, commit current copy and use sed to bulk change requires -> require
16:54 renoirb That's what I'm doing, in one commit :)
16:54 rcb on second thoguth sed 's/^\([:white:]*- \)requires:/\1require:/g'
16:54 rcb cool :)
16:55 vejdmn joined #salt
16:55 ckao joined #salt
16:55 racooper joined #salt
16:55 cro joined #salt
16:56 tseNkiN joined #salt
16:57 rcb viq, would using a pillar like """certs: \n cert1: ['name1', 'path1', 'javadir1'] \n cert2: ['name2','path2','javadir2']""" be the right direction?
16:58 toastedpenguin joined #salt
16:58 viq rcb: I'd probably go for something like http://pbot.rmdir.de/H3hzeLRQOJsT9VMIBySvSA
16:58 omame left #salt
16:59 elsmorian1 left #salt
16:59 N-Mi joined #salt
16:59 \ask is there a way to make a conditional in an .sls?  One of our states has a small part that shouldn't run on a particular server
17:00 \ask so I'd like to add something like "unless: grain['id'] == 'foo'".
17:00 scuwolf joined #salt
17:00 rojem joined #salt
17:01 viq \ask: {% if grains['id'] == 'foo' %} something {% endif %}
17:01 KoFFiE {% if grains['id'] != 'foo' %}
17:01 KoFFiE ?
17:01 viq or yeah, that
17:01 \ask viq, KoFFiE: Oh, of course. That makes sense. Thanks! :-)
17:01 KoFFiE :)
17:01 KoFFiE np
17:03 junedm left #salt
17:03 rcb viq, then name, path and java_dir will be the "args" in the key, value pair "pillar_name, args"?
17:04 cro joined #salt
17:04 vejdmn joined #salt
17:04 ckao joined #salt
17:04 racooper joined #salt
17:06 viq rcb: args["path"]
17:06 rcb viq: ok! Is this correct? http://dpaste.com/1513043/
17:07 rojem joined #salt
17:07 davet joined #salt
17:07 UtahDave joined #salt
17:07 viq sorry, gotta run
17:08 rcb viq: no prob, thnx!
17:09 rcb micw: When you get the log, check http://dpaste.com/1513044/, maybe it is closer to what you need.
17:09 ckao joined #salt
17:09 blee joined #salt
17:11 jasg75 joined #salt
17:14 \ask {% elif grains['localhost'] == 'foo' %} gives me "TypeError: 'NoneType' object is not iterable"
17:15 vejdmn joined #salt
17:15 cro joined #salt
17:15 racooper joined #salt
17:15 KyleG joined #salt
17:15 KyleG joined #salt
17:17 \ask oh, I figured it out -- it was because I made a conditional so the output ended up being "include: " but with no list.
17:18 blee_ joined #salt
17:18 jimallman joined #salt
17:21 cro joined #salt
17:22 bhosmer joined #salt
17:24 cro joined #salt
17:26 zwevans left #salt
17:27 crane joined #salt
17:27 StDiluted joined #salt
17:29 or1gb1u3 joined #salt
17:30 bemehow joined #salt
17:36 baffle joined #salt
17:36 dwfreed I'm curious, what is the purpose of the master branch?
17:36 foxx joined #salt
17:38 sine_nitore joined #salt
17:38 jalbretsen joined #salt
17:39 krak3n` dwfreed i would assume master contains the most recent "stable" released code and hotfixes which get back merged into develop
17:40 mattmtl joined #salt
17:41 jdenning joined #salt
17:41 jcsp joined #salt
17:42 stantont joined #salt
17:42 dwfreed krak3n`: seems like that would be the 0.17 branch instead? unless you mean "stable" mainline, like how Linux does -rc releases; iow, "we hit a point in develop where it's mostly stable, so that's now master; master gets hotfixes to that which are also applied in develop"
17:44 renoirb I just seen that we have now states for npm modules, only with salt 0.17.4
17:44 renoirb But I cannot upgrade to it on Ubuntu, is it normal?
17:45 timoguin master is the development branch
17:45 timoguin major releases get tagged and branched off of it
17:46 timoguin and bugfixes get cherrypicked from the master branch onto the release branches
17:48 dwfreed timoguin: that's confusing, because I thought develop was the development branch?
17:48 * timoguin goes to look
17:48 bhosmer joined #salt
17:50 timoguin well i'm not sure then. pretty sure they changed that. master being the develop branch is pretty standard for git
17:51 whiteinge the master branch was once intended to be a stable branch that saw frequent merges from develop. no on ever took that on though so it's crazy out of date
17:51 forresta renoirb, 0.17.4 was JUST released, so the ubuntu package might not be built yet
17:51 renoirb oh, ok. I was currently looking at the ubuntu and debian apt repositories to not find them.
17:51 renoirb makes sense.
17:52 psyl0n joined #salt
17:52 forresta renoirb, yea it's a double edged sword, if the guys wait to update the note saying the release is available, people complain, if they update the note and the packages aren't available, people complain, heh.
17:52 jpcw joined #salt
17:53 renoirb chicken n' egg problem
17:53 renoirb :)
17:53 forresta yep!
17:54 * renoirb warns that I'm a french speaker and you are authorized to correct my english mistakes if they irritate you
17:54 esogas_ joined #salt
17:55 renoirb An error message is unclear though:  'State npm.installed found in sls {state name} is unavailable'
17:55 renoirb That might mean that the npm.installed salt stack component is not available on the target then, right?
17:55 [diecast] joined #salt
17:56 forresta renoirb, most likely
17:56 renoirb ok, thx
17:56 forresta I thought npm was added in 0.17 for some reason
17:56 forresta but maybe the docs are incorrect, I'll take a look at the tags in a bit to see what it looks like
17:57 renoirb Me too, in the docs somewhere it said it is 0.17.4 (cannot remember where) and in http://docs.saltstack.com/ref/states/all/salt.states.npm.html says it is 0.17
17:57 dwfreed whiteinge: should probably just be killed then? from what I've seen so far, develop seems to be pretty stable, and any bugs that arise are fixed pretty darn quickly
17:58 renoirb It seems there forresta : https://github.com/saltstack/salt/blob/v0.17.2/salt/states/npm.py
17:58 brianhicks joined #salt
17:58 scoates sometimes when I `salt clientname stat.highstate` it just returns nothing (but I do think the job still runs)… any ideas?
17:58 scoates if I highstate again, it tells me that it's already running the job.
17:59 aranhoide joined #salt
18:00 scoates *state.highstate … typo in IRC, not actually an incorrect command
18:00 thrashr8_ joined #salt
18:01 forresta scoates, what happens when you run it via -l debug? Do you see additional details?
18:01 mgw joined #salt
18:01 scoates forresta: I see the initial stuff and then it just drops back to the commandline
18:01 whiteinge dwfreed: yeah, i think you're right about that
18:02 dwfreed renoirb: you need to ensure that you have the actual npm package installed on the target system
18:02 forresta scoates, and you're running this against a single host?
18:02 scoates forresta: yep
18:02 renoirb dwfreed: thanks, I just found that out too
18:02 renoirb It solved my problem too
18:02 forresta oh good point dwfreed, assumptions gooooooooo
18:03 forresta scoates, can you try to increase the timeout in your master conf, then restart the master service and try to rerun the highstate?
18:04 scoates forresta: e.g. http://paste.roguecoders.com/p/2d487919bb112aed0004db6189c1b770.txt
18:04 scoates forresta: yeah, sure. I can try that.
18:04 scoates `timeout: 5` what is a sane value?
18:04 scoates 30?
18:04 forresta just try something like 30
18:04 forresta yea
18:05 forresta just to see what happens
18:05 renoirb strange thing is that npm.installed: do not support - names:  with a - package name list
18:05 forresta but this doesn't even seem like it's getting that far..
18:06 scoates forresta: well, that time it worked. sometimes it worked with `timeout: 5`, too. I'll leave it at 30, though and see if that helps.
18:07 forresta scoates, ok cool, run it when you can with debug -l to see if something is hanging for a long time
18:08 jslatts joined #salt
18:08 scoates yeah. I am. it hangs at the same place I pasted.
18:10 forresta weird, are there any connectivity issues between the minion and master?
18:11 scoates not that we've observed. could be some packet loss, I suppose.
18:11 forresta weird
18:12 scoates maybe there was a connection problem
18:12 forresta yea seems odd..
18:12 scoates I'll keep an eye on it.
18:12 JordanRinke Morning
18:12 forresta ok cool
18:12 forresta hey
18:12 bhosmer joined #salt
18:12 carmony scoates: so now that its been a month since True North, how are you liking salt? :)
18:12 Ahlee salt-ssh doesn't use user salt runs as's key?
18:13 rcb scoates, have you tried running service salt-minion stop ; salt-minion -l debug on the client and then re-running the cmd?
18:13 scoates rcb: nope. if I can get it to happen to *my* VM (not jperras's), I'll do that.
18:14 scoates carmony: a month already? wow. such time pas. so time flying. wow.
18:14 scoates carmony: but yes; we'll be moving everything over (-:
18:14 jperras it's already fixed most issues with puppet we've had
18:15 carmony scoates: awesome, now we just need to get you to fix your emoticons to the right orientation. :P
18:15 scoates carmony: good luck. that habit is going on 20 years.
18:16 carmony scoates: thats understandable. Its about as hard to break as my nano habbit.
18:16 matanya joined #salt
18:21 forresta ...
18:21 forresta carmony, what 'nano' habit
18:21 forresta you mean the habit of being angry when you see it on a server? Because if so, that's justifiable :P
18:23 higgs001 joined #salt
18:23 MTecknology forresta: Do you know if there's any list of tags that the reactor system might see?
18:23 timoguin that's been discussed on the list recently.
18:23 forresta MTecknology, I don't, I haven't played with reactor at all :(
18:23 timoguin think the consensus is that there isn't
18:23 MTecknology minion_start apparently doesn't exist anymore and auth got changed to salt/auth
18:23 forresta hmm
18:24 whiteinge MTecknology: there is no master list of event tags yet. there's a ticket to get those documented
18:24 MTecknology It also seems like what's in 'data' may have changed
18:24 bhosmer joined #salt
18:24 MTecknology Any ideas where minion_start went?
18:25 whiteinge listen for 'salt/minion/*/start'
18:25 JordanRinke joined #salt
18:25 MTecknology Thanks!!!!!!
18:25 whiteinge MTecknology: have you seen the eventlisten.py script?
18:26 MTecknology Is there any documentation on the changes to what's in data?
18:26 MTecknology I have not
18:26 whiteinge as things stand right now, i can't write anything that uses the reactor without first consulting the output from that script
18:26 whiteinge https://github.com/saltstack/salt/blob/develop/tests/eventlisten.py
18:26 jacksontj joined #salt
18:27 timoguin very useful for seeing how certain things work in the background
18:27 linuxnewbie joined #salt
18:28 linuxnewbie left #salt
18:28 whiteinge you can just run it as root if your salt is in the normal place. if it's not:  python eventlisten.py --sock-dir /path/to/var/run/salt
18:28 MTecknology nice!
18:29 whiteinge not as nice as docs or tests...
18:29 forresta lol
18:29 whiteinge ...but it'll get you up and running now at least. :-P
18:29 MTecknology Ya- this is exactly what I've been bashing my head to try to find!
18:30 MTecknology whiteinge: I buy you drink?
18:31 aranhoide joined #salt
18:31 MTecknology not now... but in a month, I think I need to buy you one
18:34 MTecknology BAM!!!
18:34 Ahlee so, salt-ssh is effectively just wrapping calt-call in an ssh session, right?
18:34 forresta ok MTecknology, where's your doc commit, go go :P
18:34 jcockhren haha
18:35 aranhoide left #salt
18:35 cdcalef joined #salt
18:36 MTecknology forresta: I need to deal with this fire that the change caused. The commit is coming later.
18:36 forresta Psssh, if things aren't on fire, it's not a normal day
18:36 forresta just put on your asbestos pajamas in the morning
18:36 forresta problem solved.
18:37 MTecknology I'm aiming for that to change
18:37 JordanRinke any of you managing your fstab with salt?
18:37 forresta MTecknology, yea I know, I'm joking
18:38 forresta JordanRinke, as a managed file, or do you mean with http://docs.saltstack.com/ref/states/all/salt.states.mount.html ?
18:39 jslatts joined #salt
18:39 JordanRinke didnt realize that would manage fstab, assumed it was service initiated
18:39 JordanRinke (failed to read it and see the persist option - ignore my ignorance)
18:40 forresta JordanRinke, no worries
18:41 kermit joined #salt
18:41 mesmer joined #salt
18:48 andresr joined #salt
18:56 JordanRinke seriously, it is almost absurd that everything I want to do almost always has a module. Props on that.
18:57 forresta JordanRinke, yea, it's only a bummer when you found an awesome module, and there isn't a state to go along with it :P
18:58 forresta *you've
18:58 litheum joined #salt
19:00 abe_music joined #salt
19:00 JordanRinke with parted.mkpart can I leave out start/end to just have it default to the full drive like normal parted?
19:02 MTecknology -rw-rw-r--+ 1 mhansen3 0010 5951488 Dec 10 13:42 spotify.exe   <-- What the hell kind of permission is that? I've never before seen a + in there
19:03 JordanRinke acls
19:03 JordanRinke I think
19:04 JordanRinke getfacls ?
19:04 btorch MTecknology: yeah check with getfacl
19:04 druonysus joined #salt
19:05 kiorky joined #salt
19:06 JasonSwindle joined #salt
19:07 babilen joined #salt
19:07 MTecknology heh..  WARNING: unable to initialize idmapping plugin: /etc/cifs-utils/idmap-plugin: cannot open shared object file: No such file or directory
19:07 Voziv joined #salt
19:07 MTecknology btorch: JasonSwindle: Thanks! That's rather interesting
19:08 s joined #salt
19:08 jcsp joined #salt
19:08 JasonSwindle MTecknology:  What did I help with? :)
19:08 JasonSwindle I just saw your message. :)
19:09 pengunix joined #salt
19:11 MTecknology JasonSwindle: tab complete fail and your nick was close enough to JordanRinke (in shape) that I didn't notice... :P
19:12 JordanRinke Jason and I are shaped nothing like eachother!
19:12 JasonSwindle Oh, ok.   :)  JordanRinke is more awesome than I
19:12 rocket joined #salt
19:13 MTecknology J[a-z]+[A-Z][a-z]+
19:14 fxhp hey akoumjian will salt-vagrant update the version of salt on an existing box ?
19:14 fxhp (if I change the Vagrantbox version)
19:14 akoumjian fxhp: It will if you tell it to "always_install=true"
19:14 bhosmer joined #salt
19:15 akoumjian Default behavior is to detect salt, if it's installed leave it be, unless you have that option set
19:17 rocket does anyone have any advice on setting hostnames randomly etc?
19:17 jslatts joined #salt
19:20 frosty996 left #salt
19:21 Gareth 'lo
19:22 zfouts joined #salt
19:23 jcockhren rocket: I pick them from apg
19:23 cdn joined #salt
19:23 rocket jcockhren: I am not sure what you mean by apg?
19:24 jcockhren http://manpages.ubuntu.com/manpages/hardy/man1/apg.1.html
19:24 * whiteinge high-fives Gareth
19:24 whiteinge with SCaLE CFP done, it's time to start looking into travel arrangements...
19:24 Gareth :)
19:24 jcockhren I just config it for sane hostnames and run it. then pick one
19:24 rocket jcockhren: ah .. got it :) thanks
19:25 * whiteinge senses changes in seasons by SCaLE plans
19:26 KyleG I'm debating if I want to go this coming year..
19:27 KyleG It's 10 minutes away from the office/house so I probably should, might learn something. Didn't learn that much last year, but I did learn a little bit about salt that I didn't know before which was cool.
19:27 toastedpenguin joined #salt
19:28 whiteinge KyleG: you're clearly not attending the right after-parties...
19:28 KyleG :( THERE ARE AFTER PARTIES?!
19:28 JordanRinke is there a way to make a require on another id in the same state file?
19:28 ashtonian joined #salt
19:29 Gifflen joined #salt
19:31 cdn Hi guys, had this issue the other day but never managed to solve it, I was trying to use a value i'm retrieving via a grain to get a value from a pillar
19:31 whiteinge JordanRinke: yes, it takes a list so you can require multiple pkg/service/etc in the same state
19:31 cdn jdk:   pkg:     - installed     - name: {{ salt['pillar.get'](grains['test.scenario']) + "jdk.pkg") }}
19:31 rhand joined #salt
19:31 cdn where test.scenario is a grain being set on the minion
19:32 JordanRinke got it, had my order messed up
19:34 cdn had one too many ) when I typed that up. and should be '] + "jdk.pkg") }}
19:35 cdn but I get an indefined jinja variable error
19:35 whiteinge cdn: your syntax looks fine (although ~ is the official string concat operator in jinja; + can sometimes do type casting)
19:36 cdn h,mmm
19:36 whiteinge cdn: are you sure you have a key in pillar with that name?
19:37 cdn I have a key with the value of that grain in the pillar but not test.sceanrio, rather the grain value.
19:37 MTecknology Any ideas what's going on here? http://dpaste.com/1513265/
19:37 cdn Is there a way to see a more detailed error message?
19:37 MTecknology I 'thought' things were working... but apparently not
19:37 MTecknology I also need this working within about the next 10min.... damned deadlines
19:38 higgs001 joined #salt
19:39 pcurry_nomi joined #salt
19:40 whiteinge MTecknology: are you using master_tops? if not, possible syntax error in your top file?
19:41 whiteinge or perhaps whatever minion id salt-call is using doesn't match any of the patterns in your top file?
19:42 hhenkel joined #salt
19:43 jslatts joined #salt
19:43 shadfc joined #salt
19:43 blee joined #salt
19:44 JordanRinke anyone have an example of partitioning with salt? I am just randomly trying crap at this point
19:46 KyleG joined #salt
19:46 KyleG joined #salt
19:46 shadfc can i serve files out of pillar?  i'm thinking specifically for user public keys. or do I need to store those in a salt directory?
19:47 Gareth file contents in pillar?
19:48 shadfc yeah.  Just trying to figure out the right way to setup users
19:51 xmltok joined #salt
19:52 whiteinge cdn: not exactly. you can try that jinja lookup in the python shell, or you can run ``salt-call pillar.items`` on that minion to make sure the pillar value you're expecting is there. (e.g,  "<valueofgrain>.jdk.pkg")
19:53 JordanRinke being salty about it, how would yall approach needing to mount and format a drive on a vm? I can do it by hand before applying salt to them, but... it seems so... barbaric.
19:53 allanparsons joined #salt
19:53 whiteinge shadfc: look at the ``contents_pillar`` argument to file.managed
19:53 whiteinge shadfc: http://docs.saltstack.com/ref/states/all/salt.states.file.html#salt.states.file.managed
19:53 allanparsons anyone using salt on EC2?
19:54 shadfc allanparsons: i am playing around with it there, yes
19:54 shadfc whiteinge: thx
19:54 allanparsons i was wondering a) how you guys name your nodes (nomenclature)  2) if you have a script to set hostname locally + dns and 3) once you set a hostname, how do you force that cert to regenerate + re-check into salt-master?
19:54 allanparsons right now everything is ip-xxx-yyy-zzz-aa
19:56 cdn whiteinge, it works when I query the <valueofgrain> but not <valueofgrain>.jdk
19:56 whiteinge so the pillar key doesn't have the .jdk suffix?
19:57 shadfc allanparsons: i haven't figured anything out regarding targetting new nodes. I only have a master and a minion so far to play with.  I was imagining that if I were building AMI's for specific purposes (API server, etc), that I would include a static grain or something in the AMI
19:57 shadfc allanparsons: or perhaps there is a way to query for instance tags and target on those. not sure
19:57 fxhp akoumjian: ok so that worked, I got stumped for a momeny
19:58 cdn I noticed the following of a built in pillar value didn't work.   ''salt-callt pillar.items master.user''
19:58 fxhp akoumjian: "vagrant up" will not reinstall salt but "vagrant provision" will if always_install=true is configured
19:58 cdn oops, ''salt-call pillar.items master.user''  returns only     local:  ------- and no value
19:59 cdn while ''salt-call pillar.items master'' returns master and all it's children values
19:59 MTecknology whiteinge: I'm going to go with salt-call not getting the right id
20:00 whiteinge MTecknology: are you running it on your master or on one of the minions?
20:00 JordanRinke no one with the partitioning eh? I guess it is an unusual requirement.
20:00 MTecknology on a minion
20:01 whiteinge ``salt-call grains.get id`` return what you expect?
20:02 MTecknology yup... it is
20:02 cdn yup, gives me the id for my minion
20:02 MTecknology I haven't mada any changes to top.sls in a loooong time
20:03 MTecknology andI'm not using master_tops either
20:04 MTecknology with -l debug... http://dpaste.com/1513288/
20:04 whiteinge MTecknology: what does ``salt-call state.show_tops`` return?
20:05 MTecknology Function state.show_tops is not available
20:05 whiteinge what version of salt?
20:05 MTecknology uhm....
20:06 MTecknology 0.17.2 minion and 0.17.4 master
20:06 MTecknology upgrading the minion
20:06 ashtonian joined #salt
20:06 akoumjian fxhp: vagrant up wont' reinstall salt if it's coming up from a halt. If it's a fresh machine it will
20:06 * fxhp nods
20:06 fxhp learned that just now
20:07 fxhp vagrant provision will though.
20:07 MTecknology BAM!
20:08 whiteinge woot
20:08 whiteinge sad that there's an incompat there though :(
20:09 JordanRinke might want to make a note of it somewhere
20:09 MTecknology whiteinge: it's ok... it's an incompatibility with error reporting and not with actually running things
20:10 bhosmer_ joined #salt
20:12 stantont joined #salt
20:12 MTecknology I forgot that I'd been working in another sls and the pillar isn't setting the right data for it yet... error was because of that variable not existing
20:15 Sheco joined #salt
20:19 cewood joined #salt
20:20 cdn Any ideas what I can try next guys?
20:23 andresr joined #salt
20:30 jimallman joined #salt
20:38 cachedout joined #salt
20:39 bhosmer joined #salt
20:39 cewood joined #salt
20:40 whiteinge cdn: you mentioned before that it works if you leave out the ".jdk", right? so you can run: ``salt 'someminion' pillar.get <thekey>`` successfully?
20:52 Marion joined #salt
20:55 ggoZ joined #salt
20:56 cdn whiteinge
20:57 cdn That seems to work, but can't get anything like key.subkey
20:57 cdn So that might be the issue
20:58 aleszoulek cdn: It is key:subkey
20:58 cdn Just noticed it works if I use key:subkey:subsubkey
20:58 aleszoulek :)
20:58 cdn haha
20:58 cdn let me update my SLS and giver
20:59 cdn would I use the same syntax in my SLS file?
20:59 cdn - name: {{ salt['pillar.get'](grains['test.scenario']) + "jdk:pkg") }}
20:59 cedwards joined #salt
21:00 aleszoulek Both are in the context already
21:00 aleszoulek {{ grains.test.scenario }}
21:00 thrashr888 joined #salt
21:00 aleszoulek or {{ grains['test']['scenario'] }}
21:01 [diecast] joined #salt
21:01 aleszoulek Do I get it right, that in the grains, you have a 'test' dictionary (?) with a key 'scenario'? and its value determinates a key in pillar?
21:02 cdn looks like this just worked:     - name: {{ salt['pillar.get'](grains.test.scenario ~ ":jdk:pkg") }}
21:03 aleszoulek Even the pillar is in the context
21:03 cdn Is there a more "elegant' way I can format it
21:03 aleszoulek {{ pillar["%s:jdk:pkg" % grains.test.scenario] }}
21:04 aleszoulek Or maybe you need a jinja var.. Not sure you can do that in one line
21:04 aleszoulek {% set pillar_key = "%s:jdk:pkg" % grains.test.scenario %}
21:04 aleszoulek - name {{ pillar[pillar_key] }}
21:05 aleszoulek You know, both grains and pillar are variables already in a jinja context
21:05 aleszoulek When you are doing salt[X](Y)  - that is calling an X module and adding extra complexity and time to render the file
21:08 whiteinge note, you can't do the 'key:subkey:value' lookup syntax without the function call
21:11 bemehow joined #salt
21:12 cdn ahh, probably what I just did and didn't work :)
21:13 cdn I think it gives me an unknown jinja variable with I try to use pillar_key
21:14 AdamSewell joined #salt
21:14 AdamSewell joined #salt
21:15 jslatts joined #salt
21:16 btorch is possible to make pkgrepo overwrite the file ?
21:24 [diecast] joined #salt
21:24 cachedout joined #salt
21:26 `cce joined #salt
21:32 mesmer joined #salt
21:35 higgs001 joined #salt
21:35 racooper joined #salt
21:36 MicW joined #salt
21:36 MicW hi
21:36 MicW what's wrong with that file:
21:36 MicW http://pastebin.com/5UW1M4RW
21:36 MicW it never gets created and i cannot see why
21:37 shadfc you need dashes before name, source, source_hash
21:37 shadfc - name: "C:/..."
21:38 cachedout joined #salt
21:42 xmltok i know i can only run a command if another command exits in a specific condition, but the command i want to run needs a file that i only want to exist if the same specific condition is met. can i make a state depend on another state?
21:43 tonthon joined #salt
21:45 rjc joined #salt
21:46 GradysGhost joined #salt
21:46 forresta xmltok, yes
21:46 GradysGhost Hey everyone, I've got another oddity. Hopefully I can find some help here.
21:46 forresta include the state at the top of your file, then use - require: \n   - sls: asdf
21:47 GradysGhost I've got a managed file/jinja template. Running a highstate throws an error like this: Undefined jinja variable; line 331 in template
21:47 GradysGhost However, the file it references is only 119 lines long.
21:47 forresta GradysGhost, add - template: jinja in that block if it doesn't exist
21:47 GradysGhost It's there.
21:47 MicW shadfc, dach before C:?
21:47 MicW dash
21:48 cdn Is there a way to "echo" a grain so I can see it when my state is running
21:48 forresta hmmm, can you pastebin/gist your state, someone else was having this issue yesterday
21:48 GradysGhost cdn: salt 'target*' grain.get 'grain'
21:48 MicW any reason why there's no error message that tells that?
21:48 forresta MicW, when Salt compiles things together it doesn't do a syntax check where each line is checked through, you can check your yaml using the python module if you want.
21:48 shadfc MicW: dunno. i was having yaml syntax errors just get swallowed yesterday too
21:49 shadfc ah.  there you go
21:49 forresta yep, no error reporting on specific yaml issues like that :(
21:51 zfouts joined #salt
21:53 GradysGhost This is a Python file with dictionary definitions in it. I don't see anything obvious in it, but might it be possible that something syntaxy in Python is causing this parser to think it's getting a Jinja variable when it's a Python variable instead?
21:53 patrek_ joined #salt
21:54 JordanRinke that was a glorious nap, I feel so much better now.
21:56 forresta Are you working from home JordanRinke?
21:56 forresta GradysGhost, hmm, I'm not sure.
21:56 forresta it could be due to having the templating language set as jinja maybe?
21:56 forresta so it's screwing with the variables?
21:57 forresta what happens if you remove the - template line (granted it should have only worked for jinja in the first place...)
21:57 pcurry_nomi1 joined #salt
21:57 GradysGhost I'll test, forresta. It should fail still because the file references pillar data with jinja.
21:57 forresta oh does it? hmm
21:58 forresta yea no reason to remove that then since it will bomb anyways
21:58 forresta are you sure there are no syntax errors within the python file itself ?
21:58 forresta *itself?
21:59 GradysGhost I don't think so, as my devs have done some light testing. However, I have modified the file, so I'll try filling the values in manually and see where that takes me.
21:59 JordanRinke forresta: nope, I have a giant bean bag half full of stuffing in my office for my dogs, which doubles as a place to chill out and play with them / grab a nap with my dogs
21:59 forresta JordanRinke, no one gives you grief about napping?
21:59 JordanRinke forresta: Benefits of being CTO?
21:59 KyleG so proud.
21:59 scoates carmony: we just deployed our first Salt VMs to non-ops (real people (-: )
22:00 forresta apparently :P
22:00 forresta scoates niiiiiiiiiice
22:00 Gareth scoates: ops are real people too~
22:00 JordanRinke forresta: Nah here at the startup accelerator there are actually a few beds and couches and stuff, if I wasn't here all the time I might get some grief but eh ya know
22:00 scoates Gareth: I should have said "normal people" (-;
22:00 forresta JordanRinke, yea makes sense.
22:01 bhosmer joined #salt
22:01 pengunix joined #salt
22:01 pcurry_nomi joined #salt
22:02 MicW can i somehow write use %TEMP% in a path?
22:02 GradysGhost forresta: No, no syntax errors.
22:02 forresta GradysGhost, odd, what does the traceback actually say?
22:02 forresta other than there being an error on a line that doesn't exist
22:03 GradysGhost [DEBUG   ] Jinja search path: '['/var/cache/salt/minion/files/base']'
22:03 GradysGhost [ERROR   ] Undefined jinja variable; line 331 in template
22:03 GradysGhost And then it prints out that template file.
22:03 GradysGhost That's all I get, and that's with -ldebug
22:03 GradysGhost Would ramping that up to -lall give more/better data?
22:04 forresta maybe, but I don't it would help in this case, you could try it though
22:05 GradysGhost I just did, no additional data.
22:05 bhosmer_ joined #salt
22:06 forresta ok, can you try removing ALL jinja variables from the script, then see if salt runs the managed file then?
22:06 GradysGhost Sure, I'll comment those lines out
22:07 GradysGhost Leave the " - template: jinja" bit in place?
22:07 forresta yes, let's try one change at a time
22:07 carmony scoates: awesome!
22:07 forresta GradysGhost, if it still errors afterwards, we can remove that
22:07 jimallman joined #salt
22:07 forresta carmony, you never answered the nano question :P
22:07 scoates carmony: yep. thanks again.
22:08 carmony forresta: I plead the 5th :)
22:08 forresta lol
22:08 scoates Managing my users' `Vagrantfile`s via a Jinja template and a file.managed on `/vagrant/Vagrantfile` … pretty great. (-:
22:08 scoates we used to have a PHP app that generated Ruby (Vagrantfile). Ugly. (-:
22:09 forresta bleh
22:09 carmony forresta: since I'm pretty much a self-taught guy, when I was very first learning linux the tutorial used "nano" and I didn't have anyone to tell me the benefits of using vi, so it just became a habit to use nano, though now I do know how to use vi
22:09 forresta so scoates, when are you gonna put this project up on github so I can add it to the projects page?
22:09 GradysGhost Handles it just fine.
22:09 forresta carmony, yea I was just joking around giving you grief.
22:09 forresta GradysGhost, so no errors?
22:09 carmony forresta: oh I know :)
22:09 forresta ok good
22:09 scoates forresta: we're at least a month from release. This was for dev VMs only for right now.
22:09 GradysGhost No errors. 1 successful change.
22:09 forresta scoates, gotcha
22:09 JordanRinke On cmd.run the unless and onlyif states, will those return on non-zero exit codes? like if I run a command and it exits 0, an onlyif would pass, and an unless would fail? and any non-zero error would make an unless pass?
22:10 forresta GradysGhost, interesting....
22:10 scoates I do intend to port rushmore.fm over the holidays, though.
22:11 forresta GradysGhost, ok next can you try adding a simple call to a locally defined item with jinja, so just a {% set test = 'testvalue' %}
22:12 forresta and then see if that works, then on the next test, actually use the {{ test }} variable somewhere
22:12 GradysGhost Sure. I was going to start adding those things back in one at a time to see if one of them is breaking it.
22:12 forresta awesome
22:14 forresta JordanRinke, unless basically says 'if the thing I am watching fails, run this other thing', onlyif is 'if the thing I am watching passes, run this other thing'.
22:14 GradysGhost The test you propose is successful. It fails when I add one of mine back in. Guess something's wrong with that.
22:14 JordanRinke yeah, is that success/fail critiera based on a 0 exit code?
22:14 forresta GradysGhost, sweet
22:14 Hefeweizen joined #salt
22:14 GradysGhost We're on the right track. Lemme take a closer look at this.
22:14 forresta JordanRinke, I believe so, but I don't remember off hand :\
22:14 carmony scoates: how is everyone at Fictive Kin enjoying salt now?
22:14 forresta GradysGhost, awesome, let me know what you find
22:15 GradysGhost Would single quotes screw up Jinja?
22:16 GradysGhost Like '{{ some_var }}' ?
22:16 scoates carmony: just one project
22:16 GradysGhost Hah! I think I just found the problem. One sec...
22:17 scoates carmony: and it's better if they don't even know (-: but jperras and I are enjoying it, yes.
22:17 cdn is there anyway from within a state to output to the screen when it runs, a grain value?   Don't want to do it via command line, just to use it sorta like progress
22:17 carmony nice :)
22:18 cachedout joined #salt
22:18 GradysGhost I think it may have something to do with quotes. I need the variable to be surrounded by quotes so Python recognizes it as a string, but I also need to use quotes to specify the Jinja variable. Testing...
22:19 forresta GradysGhost: http://jinja.pocoo.org/docs/templates/#literals
22:19 forresta perhaps try double quotes?
22:19 forresta that example there has a jinja variable wrapped in double quotes no problem
22:19 JordanRinke forresta: tested, appears to be the case. onlyif = 0 exit code, unless = any exit code non-zero
22:20 kaoskitn left #salt
22:20 forresta JordanRinke, ok cool, so expected behavior then, thanks for the heads up!
22:20 GradysGhost I also think this has something to do with my YAML-fu being weaker than my JSON-fu.
22:20 forresta well, start writing your states in json :P
22:21 forresta but I don't see any examples of wrapping in single quotes other than an example of an escape character on the jinja docs :\
22:21 jcockhren forresta: get on that
22:21 forresta no way man, the jinja docs are awesome
22:22 jfzhu_us joined #salt
22:22 forresta and I have enough issues I've opened that I need to resolve myself :P
22:24 forresta hey basepi, if you see techhat around today, could you ask him if he has an ideas for: https://github.com/saltstack/salt/issues/9319 ?
22:24 forresta if he isn't super busy that is
22:25 basepi forresta: he and i are both working from home
22:25 basepi major snow storm here
22:25 forresta oh ok
22:25 basepi freezing rain, closed airport, the whole 9 yards
22:25 basepi i'll see if i can ping him and have him ping you, though
22:25 forresta I'm looking out the window at green trees, and had to take my hoodie off on the way into the office
22:25 basepi hahahaha
22:25 forresta basepi, nah it's not that big a deal, only been 2 days.
22:26 forresta if he hasn't responded in a week I'll bug you again
22:26 GradysGhost Okay, so my pillar is some DB connection options (host, port, schema). Here's a sanitized version of what it looks like: http://www.hastebin.com/winadugure.sm
22:26 GradysGhost oh crap
22:27 GradysGhost my god, I'm going to feel so stupid if I just figured this thing out and it is what I think it is.
22:27 forresta GradysGhost, ?
22:27 GradysGhost I'm not accessing it with {{ pillar['yadda']['yadda'] }} but with {{ yadda['yadda'] }}
22:28 GradysGhost It's been a long day.
22:28 forresta lol
22:28 forresta did that fix it?
22:28 forresta It still annoys me though that it reports there is a jinja issue on a line that doesn't exist. I don't understand why that happens :\
22:29 MK_FG joined #salt
22:30 scoates I've run into that. It's frustrating.
22:30 forresta scoates, yea I don't get what causes it either
22:30 elithrar joined #salt
22:32 GradysGhost forresta: No, not fixed, same errer, with a reasonable line number. I'm checking pillar data now to make sure it's being delivered. I believe that it is, though.
22:32 elithrar Is there a way to set your hostname on Debian systems with Salt (without just regexing /etc/hostname and /etc/hosts ?)
22:32 forresta ok, so at least that fix provided some more hints on where to go, cool
22:33 basepi forresta: apparently joseph just finished writing that, and will be committing it in the next few minutes.
22:33 forresta oh the example?
22:33 forresta sick
22:33 scoates elithrar: see /etc/init.d/hostname.sh
22:33 JordanRinke ssooo dirty
22:34 forresta lol
22:34 JordanRinke - unless: sfdisk --print-id /dev/sdb 1
22:34 JordanRinke - name: parted -s -a optimal /dev/sdb mklabel gpt -- mkpart primary ext4 1 -1
22:34 forresta ugh
22:34 forresta why
22:34 KyleG what about /etc/sysconfig/network ?
22:34 zfouts joined #salt
22:35 forresta elithrar, https://github.com/saltstack/salt/issues/4390 that was opened a long while back, looks like Tom is suggesting to use a runner: http://docs.saltstack.com/ref/runners/index.html
22:35 GradysGhost BAM!
22:35 GradysGhost Got it.
22:35 forresta what did it end up being?
22:35 GradysGhost Thanks, everyone, for bearing with me as I facepalmed everywhere.
22:35 GradysGhost Syntax.
22:35 blee_ joined #salt
22:35 forresta past the pillar issue?
22:35 GradysGhost Well, I don't know exactly why it works now when it didn't work 30 seconds ago, but I know why it didn't work before that.
22:35 elithrar forresta: Great, thanks!
22:36 forresta elithrar, yea np, I don't know if a runner exists to do that from the premade ones though, you might have to write one
22:36 GradysGhost Before that, I was not using pillar['whatever']
22:36 GradysGhost Though I knew better.
22:36 forresta ahh so same issue as above
22:36 GradysGhost yep
22:36 GradysGhost Thanks, man.
22:36 forresta Yea np
22:37 elithrar Worst case is that I'll do it with Vagrant/Packer (building machine images) but would be nice to keep it self-contained
22:37 forresta elithrar, yea for sure
22:37 forresta elithrar, if you end up having to write the runner, can you see about making a pull request to get it back into Salt?
22:39 forresta basepi, if you guys were in the office I'd say techhat was deserving of a high five.
22:39 dstanek joined #salt
22:40 basepi give him a virtual one on the issue tracker!  =P
22:40 forresta lol
22:42 forresta how is there no high five emoji
22:45 rgarcia_ joined #salt
22:45 wasd_ joined #salt
22:46 rgarcia_ is there a way to do the equivalent of "ssh -v" with salt-ssh? getting permission denied errors and would like to see if it's using the private key I specified
22:48 pcurry_nomi joined #salt
22:50 JordanRinke oh man, this is so nasty I almost like it
22:51 forresta rgarcia_, I checked http://docs.saltstack.com/ref/cli/salt-ssh.html and I don't see an option to do so.
22:51 forresta rgarcia_, if you can't find a workaround for it, create an issue on the tracker to see if that can be added.
22:52 rgarcia_ forresta: thank
22:52 rgarcia_ thanks*
22:52 forresta rgarcia_, yea np
22:52 JordanRinke forresta: how do I require something else in the same state file?
22:52 JordanRinke like I have an id partition_sdb, and I want format_sdb to require partition_sdb in the same state file
22:52 forresta - require: \n  - thing: name
22:53 forresta are they part of their own 'blocks'?
22:53 JordanRinke right, what is 'thing' in this case?
22:53 JordanRinke http://pastebin.com/gjFGi2Y4
22:53 forresta whatever it might be, so if partition_sdb was a pkg, you'd use pkg
22:53 forresta use - require: \n  - cmd: partition_sdb
22:53 JordanRinke ah cmd
22:53 JordanRinke thx
22:53 forresta np
22:54 JordanRinke ohhh because the block name inherits the type of action it is?
22:54 forresta right
22:54 JordanRinke or, is defined as that more accurately I guess
22:54 JordanRinke ok, that makes sense, I was thinking of it as it's own thing, not as a named whatever
22:55 forresta ahh gotcha
22:55 JordanRinke bwahaha that works, oh... that is... wonderful
22:55 Ryan_Lane joined #salt
22:58 elithrar joined #salt
22:59 bemehow joined #salt
23:05 jimallman joined #salt
23:13 bhosmer joined #salt
23:22 alunduil joined #salt
23:32 bemehow joined #salt
23:38 bemehow joined #salt
23:44 mesmer joined #salt
23:48 JordanRinke is there a way to background a highstate? or well, let me explain...
23:48 JordanRinke I have xtradb setup, and my state sets it up/joins a cluster but then is has to replicate 20GB of DB data, the salt command hangs waiting for the service start to finish
23:48 JordanRinke which hangs, waiting for the data to replicate, make sense?
23:49 scoates JordanRinke: --async ?
23:50 JordanRinke my google fu is weak
23:50 JordanRinke thx
23:53 pcurry_nomi1 joined #salt
23:57 jacksontj joined #salt
23:58 rw4 joined #salt

| Channels | #salt index | Today | | Search | Google Search | Plain-Text | summary